Ara Wagoner is from the Android Central.
Cloud service disruptions are not new. The shift to working from home in 2020 exposed tons of vulnerabilities, as carriers, cable and fiber companies, and every popular app under the sun experienced some temporary, catastrophic collapse. It placed an unprecedented burden on the cloud infrastructure systems that support your favorite sites. The consequences of these outages were inevitable.
It would have been nice to see marked improvement in 2021. The internet is a deck of cards ready to collapse if the wrong piece is folded. When a site puts all their data and traffic in a single cloud basket, just one failure can take out some of the highest-traffic sites, when we'd expect these sites to have better contingencies in place.
Our favorite messaging apps, smart homes, gaming networks, productivity suites, and social media sites all collapsed this year. Beyond that, the Amazon Web Services (AWS) and Facebook outages proved how much of our daily lives depend on the cloud.
We can only hope that things will improve in 2022. Unless cloud infrastructure companies and content delivery networks change the way they do things, there is no reason to assume they will.
1. The Amazon Web Services outage stopped deliveries, cameras, and cat feeders.
The recent December outage is still fresh in your mind. When Amazon Web Services fell apart, it may have taken a third of the cloud services with it.
A surge of connection activity that overwhelmed the networking devices between the internal network and the main network resulted in delays for communication between these networks. Because the internal network is linked to the globalAWS server, it caused traffic delays or site shutdowns for about 7 hours until the devs could fix it.
During the holiday shopping season, Amazon delivery drivers' apps went down, leaving them unable to complete deliveries. Companies missed out on revenue because consumers couldn't make new Amazon orders. The smart video doorbells and baby monitors that were part of the Prime Video and Music services were temporarily worthless. Disney+, Venmo, and iRobot all broke down because of their choice of cloud provider.
According to CNBC, the effects of the Amazon.com outage rippled out to disrupt final exams at colleges, since some exam services relied on the cloud to work. Some automatic cat feeders stopped feeding their cats.
Readers of the Android Central said they were more warier about cloud- dependent smart home tech after this outage. Experts think it's unlikely that Amazon will incorporate offline controls into its smart home tech. The cloud allows them to sell cheap, underpowered tech that wouldn't work without it.
2. The Meta-verse is broken.
Nick Sutrich is from Android Central.
We have to mention Facebook if we're talking the messiest 2021. Facebook accidentally shut down its own cloud services due to "configuration changes on the back end that coordinate network traffic between our data centers" which cascaded and brought down all of their online services. It made sure no one could access Meta services, including its own employees.
Even though Meta's cloud server only power its own businesses, this outage still hurt other companies. Meta's server and token shut down as well as any sites that relied on Facebook logins, while other sites that relied on Meta's server or token shut down as well.
The cloud-powered peripherals were undermined by the Facebook outage. Due to the requirement of a Facebook account, the library of games for Quest 2 owners could not be accessed. We said at the time that Facebook needs to add offline support for its tech in the future.
The worst fiasco for the company was the 6-hour Whatsapp outage. For millions of people who use the app as their primary way to communicate, a single day without it was too many. 70 million new members joined Telegram after the outage. That doesn't mean that it lost a lot of users, but it definitely saw a significant exodus that it might never recover from.
In April 2021, there was a similar outage that lasted 45 minutes.
3. The internet is fastly taken down.
You don't pay attention to things that work. Fastly's content delivery network broke in June, dragging down some of the most popular websites with it.
Many companies rely on the help of a CDN to cache their content for faster loading times and reduced bandwidth load. Load times are kept low regardless of where the user lives because they deliver data at high speeds across the globe.
A faulty service configuration caused disruptions across our POPs globally, which hurt the sites that relied on its edge computing. In early June, a number of websites went at once, including Amazon, The New York Times, and CNN. This was a broad but relatively short-lived outage compared to the rest, and it was restored within 49 minutes.
4. There were four PSN outages for a messy PS5 year.
The source of the information is the Android Central.
If you've got a PS5 this year, you're probably going to run into problems accessing your library or playing games in the future. Sony and Akamai have had several issues this year.
Some PS5 and PS4 players couldn't access their gaming libraries for several days during the worst of the PS5 and PS4 outage.
Sony had network issues that needed to be worked through. In each case, players around the world would get error messages when they tried to access online services, with the outages lasting from 1 to 5 hours.
Many of the best PS5 games require constant online connections. If Sony can't keep its service up and running for days at a time, that will make its fans unhappy.
5. The company can't help its customers with their smart Home.
The first major outage of the year was in February. If you tried to ask a question about the device, you would be told it was not yet set up. It was impossible to connect to your devices associated with your account because of that. There were issues with the Google Assistant app.
The impact on all users of the Home was apparent that evening, with users taking to the forums for help. It's not clear when the issue started, but it was fixed by the company that evening.
6. The smart home winks out.
The source is Wink.
Most of the worst outages affected a wide range of sites. The worst outage of the year was by Wink Hubs, which shut down for 10 days. The hubs were no longer able to control the products because of their dependence on cloud services.
As an apology, Wink offered a 25% discount on its subscription costs, but we don't know what caused the issue, only that it would "optimize the Wink Backend and ourAPI now that it is back up." Many customers saw this as a sign that it was time to abandon Wink.
7. The notifications system for the phone is kaput.
Alex Dobie is the source of the Android Central.
Delay in knowing your condition can lead to further spread and sickness. The glitch in the NHS COVID-19 app was a bad look for the company.
The screen people were looking at was an "Loading" screen. After about 12 hours of reports, it was announced that the issue would be looked into by the company. There were many reasons to distrust the app by that point, including the "phantom notification" glitch from 2020 that would pop up, then disappear before you could tap on it.
8. There was an outage of the Amazon Web Services.
The source of the information is the Android Central.
There was a second Amazon Web Services outage on December 15. This time it took out many more.
There were three Amazon Web Services outages on December 22 that shut down Imgur. The issue was caused by a power outage at an east-coast facility. That caused three outages in three weeks. The last two were only for an hour or so, which is long enough to cause problems.
Will the problem diminish or grow in the future?
Our current cloud dependent system can be fragile. A single crisis can cripple our productivity or render our expensive tech useless, because so much of our internet use concentrated on a few apps and services.
Can we hope for fewer accidents next year?
The source is the internet search engine, Google.
We need to invest more in cloud infrastructure to see fewer outages. The recent infrastructure bill has billions allotted for improving high-speed, rural broadband access and civilian cybersecurity, but most of the worst outages came from company errors, not hostile actors. We may have to rely on companies to invest in their own cloud infrastructure.
According to a report by the research firm, companies will spend $482 billion on cloud services in 2022, a 21.7% increase. That should be a step in the right direction.
It's important to note that the majority of the worst outages were caused by companies' internal monitoring networks or third-party CDNs. Human error can bring down the system in the wrong circumstances, where it can have disproportionate consequences. One more potential step where something can go wrong is where the CDNs add.
It doesn't matter how much you invest when a single server can topple the system. We need companies to structure their data better so backups can kick in quickly if there is a problem. We are in better shape than we were two years ago, but we still have a long way to go.
We might earn a commission for using our links. You can learn more.
